TO DOWNLOAD A COPY OF THIS POSTER, VISIT WWW.WATERS.COM/POSTERS ©2015 Waters Corporation

INTRODUCTION

Antibody drug conjugates (ADCs) is a sub-class of

biotherapeutics which consists of monoclonal

antibodies (mAbs) and cytotoxic drugs linked to mAbs

by chemical linkers. The reactivity of primary amines

and their compositional availability in monoclonal

antibodies (mAbs) make them a popular target for

chemical conjugation in antibody-drug conjugates

(ADCs). However, conjugation reactions of lysine

residues result in highly heterogeneous mixtures with

drugs in many combinations at different lysine sites

on the mAb. Hence, the structural complexity and

intrinsic heterogeneity of lysine-conjugated ADCs

impose a prominent analytical challenge to current

characterization methods. Quantification of

conjugated peptides and site occupancy ratio

determination was traditionally done by UV methods.

The drawbacks for UV quantification include low

sensitivity, insufficient selectivity and relative long

analysis time. MS based quantification can provide

higher selectivity and sensitivity compared to UV

based methods. The biopharmaceutical industry lacks

a complete workflow that enables efficient

identification and quantification of ADC peptides,

therefore facing a great challenge to make direct

comparison on the site occupancy of ADCs from

multiple sources. In this study, we present an

integrated approach that combines multiplexed MS/

MS data acquisition strategy with multi enzyme

digestion for the in-depth characterization of lysine-

conjugated ADCs. Both data-independent acquisition

and data-dependent acquisition (DDA) methods were

used to identify the lysine-conjugated peptides,

confirm the conjugation sites, determine the relative

site occupancy ratio and compare conjugated peptide

levels across samples.

IN-DEPTH CHARACTERIZATION OF LYSINE-CONJUGATED ANTIBODY-DRUG CONJUGATES (ADCS)

BY LC/MS QUALITATIVE AND QUANTITATIVE ANALYSIS

Liuxi Chen, Henry Shion, Ying Qing Yu, Weibin Chen Waters Corporation, Milford, MA 01757, USA

METHODS

Sample Preparation: Peptide mapping analysis: Trastuzumab (Tmab) and trastuzumab emtansine (Tmab-DM1) were denatured, alkylated and digested by trypsin and Asp-N endoproteinase, respectively.

Leucine enkephalin (LeuEnk) was added to each sample at final concentration of 50 fmol/ul as an internal standard. Enzyme digests of Tmab and Tmab-DM1 were analyzed in triplicate injections. Intact mass analysis: All samples were treated with PNGase F overnight to remove the N-linked glycans in order to reduce the intact mass spectra complexity. LC/MS: LC: ACQUTIY UPLC H-Class System Column: ACQUITY UPLC BEH C18, 300Å, 1.7 μm, 2.1 x 100 mm (p/n 186003686) Column temperature: 65 oC Mobile phase: A. 0.1% Formic acid in water

B. 0.1% Formic acid in acetonitrile

MS: Xevo G2-XS QTof MS Data Acquisition: MSE and FastDDA Mode: ESI positive mode Capillary Voltage: 3.0 kV Cone Voltage: 30 V Source Temperature:100 °C Desolvation Temperature: 250 °C Mass Range (m/z): 100-2000 Lock mass used: 100 fmol/ul of Glu-Fibrinopeptide B in ([M+2H]2+, 785.8426)

MSE settings: Scan rate for alternating low/high Energy: 0.5 sec Low energy: 6 V

High energy ramp: 20-45 V FastDDA settings: MS scan time: 0.2 sec MSMS scat time: 0.2 sec Peak detection: Intensity threshold Max. # MSMS scans/survey: 5 Dynamic peak exclusion: Acquire and then exclude for 8 sec (± 1.1 Da) Collision Energy: m/z dependent ramp applied for low and high mass Stop MS/MS criteria: TIC 5e8 or 0.4 sec Informatics: UNIFI 1.8

Intact Mass workflow Peptide mapping (MSE) workflow

Peptide mapping (FastDDA) workflow Accurate mass screening workflow UNIFI scientific library

Figure 1. Surface exposed lysine residues on IgG1.

RESULTS AND DISCUSSION

Intact Mass Analysis

ADC Peptide Analysis Workflow

The distribution of the drug load is determined by MS

intact analysis. The deconvoluted mass spectra con-tain 8 major peaks with mass difference of 957 Da

between adjacent peaks, which is in agreement to the mass of covalently linked DM1 drug with one

MCC linker. In both the innovator and candidate biosimilar ADCs, 8 major peaks correspond to Tmab

with 0-7 DM1 drug and linkers respectively (label as +0 drug, +1 drug, etc). The less abundant peaks

right next to the major peaks with 219 Da, which at-tributes to the unreacted linkers that modified the

antibody but do not react with DM1.

Figure 2. Combined raw spectra for Tmab(A)

and Tmab-DM1 (B).

Figure 3. Deconvoluted spectra for Tmab(A)

and Tmab-DM1 (B).

(A) Tmab

(B) Tmab-DM1

(A) Tmab

(B) Tmab-DM1

Figure 4. ADC Peptide level analysis identifi-

cation and the quantification workflows. Ly-sine-conjugated ADC and unconjugated con-

trol mAb were digested by trypsin and Asp N respectively, followed by MSE and DDA

modes. UNIFI 1.8 Peptide Mapping workflow was used to identify the conjugated peptides

and pinpoint the conjugation sites. The same set of MSE data were further analyzed using

UNIFI 1.8 Accurate Mass Screening workflow to quantify the relative site occupancy and

relative abundance of conjugated peptides across different samples. (US patent pend-

ing)

Peptide Mapping—Site Identification Peptide Mapping—Site Quantification

Figure 5. LC/MSE chromatogram (BPI) of tryptic peptide mapping analysis

for Tmab vs Tmab-DM1 in UNIFI comparison mode.

Figure 6. LC/MSE chromatogram (BPI) of Asp-N peptide mapping analysis

for Tmab vs Tmab-DM1 in UNIFI comparison mode.

Tmab-DM1

Tmab

Tmab-DM1

Tmab

10x

10x

Table 2. Numbers of conjugation sites identified in different regions of

Tmab using DDA and MSE methods.

Figure 8. Relative abundance of conjugated peptides (tryptic) across sam-

ples.

CONCLUSIONS

Table 1. The enzyme of choice for different

quantification purposes. Trypsin digest was used to calculate relative abundance of con-

jugated peptides, while Asp-N digest was used to determine the relative site occu-

pancy of individual site. (US patent pending)

Figure 7. MS/MS spectra to confirm conjugations sites for positional isomers

for Asp-N peptide 224 DKTHTCPPCPAPELLGGPSVFLFPPKPK251

The CID fragmentation of the

Tmab-DM1 generates a signa-ture fragment ion (m/z 547.2,

charge +1), commonly for all conjugated peptides. The signa-

ture fragment ion corresponds to a partial drug fragment.

Figure 9. Relative site occupancy determined using Asp-N digestion.

Relative site occupancy

ratio = Area (conjugated pep. peak)/

[Area(unconjugated pep.Peak)+Area

(conjugated pep. Peak)]

For Tmab-DM1, 80 out of 92 conjugation sites were observed.

UNIFI provided automated workflow for:

In-depth primary structure characterization of lysine-conjugated

ADC.

Site specific localization of ADC conjugation (Peptide Mapping

Workflow).

Quantification of relative site occupancy (Accurate Mass Screen-

ing Workflow).

While this presentation has focused on lysine-conjugated ADCs,

these UNIFI workflows are directly applicable to other classes of

ADC biotherapeutics.

References:

1. Kim MT, Chen Y, Marhoul J, Jacobson F. Bioconjug Chem. 2014 Jul 16;25(7):1223-32

2. Wang L, Amphlett G, Blattler WA, Lambert JM, Zhang W, Protein Sci.

2005 Sep;14(9):2436-46